Flux
Figma for Agents: How Airflow's Creator Coordinates AI ft. Maxime Beauchemin
Programmation IA

Figma for Agents: How Airflow's Creator Coordinates AI ft. Maxime Beauchemin

It’s hard to keep up with the AI evolution; new AI tools drop every week, but how are experienced practitioners actually using them? Most of us are overwhelmed and unsure about the many possibilities, yet we need to keep going and do our work. You might use AI agents all day long, parallelize them with AI Orchestrators, tmux, git worktree, and so on, using AI IDEs, but in the end, you still need to coordinate and understand what the agents produced, potentially test it, which makes it even…

SSP Data Engineering Blog
Ruby's Bundler 4.0.18 Extends Cooldown to bundle lock and bundle cache
Cybersécurité Programmation

Ruby's Bundler 4.0.18 Extends Cooldown to bundle lock and bundle cache

Bundler 4.0.18 adds the opt-in --cooldown flag to bundle lock and bundle cache, closing gaps where the setting could not be applied. Bundler 4.0.18, released August 5, 2026 alongside RubyGems 4.0.18, extends the cooldown feature to two more commands. The opt-in --cooldown flag now works with bundle lock and bundle cache, which previously resolved dependencies without accepting the flag. RubyGems lists the change under Bundler security in the release notes. The cooldown feature is not new. It…

Socket
An AI model from Meta also hacked another company during testing
IA Programmation

An AI model from Meta also hacked another company during testing

An AI model from Meta also hacked another company during testing Stop me if you've heard this one before: An AI model from the parent company of Facebook and Instagram hacked into another company’s systems during cybersecurity testing, a spokesperson confirmed on Wednesday. Meta says the breach occurred because of an inadvertent error during testing of the model, similar to previously disclosed incidents with OpenAI and Anthropic. “A misconfiguration by Irregular, an independent testing company…

Simon Willison's Weblog
An AI model from Meta also hacked another company during testing
IA Programmation

An AI model from Meta also hacked another company during testing

An AI model from Meta also hacked another company during testing Stop me if you've heard this one before: An AI model from the parent company of Facebook and Instagram hacked into another company’s systems during cybersecurity testing, a spokesperson confirmed on Wednesday. Meta says the breach occurred because of an inadvertent error during testing of the model, similar to previously disclosed incidents with OpenAI and Anthropic. “A misconfiguration by Irregular, an independent testing company…

Simon Willison's Weblog
Introducing Muse Code and Muse Spark 1.2
IA Programmation

Introducing Muse Code and Muse Spark 1.2

Introducing Muse Code and Muse Spark 1.2 Yet more evidence that the most important characteristic of any model these days is long-sequence agentic tool calling. Meta shipped their own coding agent as part of getting that to work! Muse Spark 1.2 is a coding-focused update to Muse Spark 1.1, with improvements in code generation, complex debugging, codebase understanding, and end-to-end developer workflows. In Muse Spark 1.2, we significantly scaled up training compute on coding tasks while…

Simon Willison's Weblog
Introducing Muse Code and Muse Spark 1.2
IA Programmation

Introducing Muse Code and Muse Spark 1.2

Introducing Muse Code and Muse Spark 1.2 Yet more evidence that the most important characteristic of any model these days is long-sequence agentic tool calling. Meta shipped their own coding agent as part of getting that to work! Muse Spark 1.2 is a coding-focused update to Muse Spark 1.1, with improvements in code generation, complex debugging, codebase understanding, and end-to-end developer workflows. In Muse Spark 1.2, we significantly scaled up training compute on coding tasks while…

Simon Willison's Weblog
Third-party cyber evaluations involving OpenAI models
IA Programmation

Third-party cyber evaluations involving OpenAI models

Third-party cyber evaluations involving OpenAI models And another one. I had to create a accidental-cyberattacks tag to keep track of them all! This post from OpenAI covers both the UK AI Safety Institute attack (see my previous post) and another attack enabled by Irregular: Irregular, one of our external cybersecurity testing partners, was running Capture-the-Flag-style evaluations intended to be isolated from the internet, but a testing-environment misconfiguration allowed models to access…

Simon Willison's Weblog
Third-party cyber evaluations involving OpenAI models
IA Programmation

Third-party cyber evaluations involving OpenAI models

Third-party cyber evaluations involving OpenAI models And another one. I had to create a accidental-cyberattacks tag to keep track of them all! This post from OpenAI covers both the UK AI Safety Institute attack (see my previous post) and another attack enabled by Irregular: Irregular, one of our external cybersecurity testing partners, was running Capture-the-Flag-style evaluations intended to be isolated from the internet, but a testing-environment misconfiguration allowed models to access…

Simon Willison's Weblog
Incident Report: unsanctioned agent behaviour during cyber testing
IA Programmation

Incident Report: unsanctioned agent behaviour during cyber testing

Incident Report: unsanctioned agent behaviour during cyber testing It happened again. This time it was the UK government's AI Security Institute who accidentally attacked other companies while running an evaluation with models with the safety filters turned off. From their technical paper (PDF): During a cyber evaluation, from 25 to 28 July 2026, AI agents engaged in sustained, unsanctioned activity directed at what were, in practice, real people and organisations. These attempts were…

Simon Willison's Weblog
Incident Report: unsanctioned agent behaviour during cyber testing
IA Programmation

Incident Report: unsanctioned agent behaviour during cyber testing

Incident Report: unsanctioned agent behaviour during cyber testing It happened again. This time it was the UK government's AI Security Institute who accidentally attacked other companies while running an evaluation with models with the safety filters turned off. From their technical paper (PDF): During a cyber evaluation, from 25 to 28 July 2026, AI agents engaged in sustained, unsanctioned activity directed at what were, in practice, real people and organisations. These attempts were…

Simon Willison's Weblog
One-shotting a Raccoon Heist game using Claude Fable 5
IA Programmation

One-shotting a Raccoon Heist game using Claude Fable 5

Back in 2024 I tweeted screenshots of a game concept generated by GPT-3 and some concept "art" created using DALL-E. Today, on the fourth anniversary of that tweet, I decided to see if Claude Fable 5 (running in Claude Code for web) could build the entire game from the content of that tweet. It did a pretty good job of it! You can play the game here. Here's the GitHub repo, and a short video demo: Your browser does not support HTML5 video. How I built this This is the August 5th, 2022 tweet: My…

Simon Willison's Weblog
One-shotting a Raccoon Heist game using Claude Fable 5
IA Programmation

One-shotting a Raccoon Heist game using Claude Fable 5

Back in 2022 I tweeted screenshots of a game concept generated by GPT-3 and some concept "art" created using DALL-E. Today, on the fourth anniversary of that tweet, I decided to see if Claude Fable 5 (running in Claude Code for web) could build the entire game from the content of that tweet. It did a pretty good job of it! You can play the game here. Here's the GitHub repo, and a short video demo: Your browser does not support HTML5 video. How I built this This is the August 5th, 2022 tweet: My…

Simon Willison's Weblog
UK Cyber Test: AI Agent Attempted to Social Engineer Open Source Maintainer Into Merging Malware
Cybersécurité Programmation

UK Cyber Test: AI Agent Attempted to Social Engineer Open Source Maintainer Into Merging Malware

An AI agent powered by Anthropic’s Mythos 5 created a malicious pull request, fabricated identities, targeted open source maintainers, and planted instructions for other coding agents during a UK government cybersecurity evaluation. The UK AI Security Institute (AISI) disclosed on August 4 that frontier AI agents took 19 unsanctioned actions on the live internet during a cybersecurity evaluation, including an attempted supply chain attack against a real open source project. The most serious run…

Socket